What’s Driving the Rise of International Star Firms?

A large academic and policy debate has focused on the increase in market concentration over the past few decades giving rise to “star firms,” a small set of firms that generate abnormal returns for their investors. A common concern is that these firms exert excess market power and behave as traditional monopolists, restricting competition. In our earlier research studying US firms (see here and here), we found little evidence of this, suggesting increases in profits were mostly due to greater efficiency and greater investment in the new knowledge economy. However, less is known about characteristics of star firms worldwide and whether they exploit their market power in traditional ways by cutting output and investment compared to other firms. Importantly, when we broaden the horizon to look internationally, the role of competition becomes unclear. Differences in firm performance may occur for a variety of reasons including variances in income level and level of financial development, trade barriers, human capital, and overall institutions. Whether star firms in emerging markets are a manifestation of greater efficiency and investment in intangible capital or a reflection of poor-quality institutions that protect incumbents and encourage anticompetitive behaviors is an open question.

In our recent paper, we take a close look at star firms around the world and how they may differ between developed and developing countries and across industries. We use micro panel data from over 70 countries over the period 1980-2017, and begin by analyzing the institutional and industry characteristics associated with the existence of star firms, focusing in particular on the role of intangible capital and competition. We define star firms as firms in the top 10 percent of return on invested capital (ROIC) in the world in a particular year.

Where do star firms occur?

Our summary statistics show that while world stars, in a given year, are more likely to occur in high-income countries and in the manufacturing sector, an increasing share are from middle-income countries and the services sector over time. For example, the percentage of stars from middle-income countries has gone from 1 percent to 47 percent during the sample period. In addition, nearly 40 percent of world stars are high intangible capital firms over our sample period, consistent with the rise of a new knowledge economy.

We also see that while star firms are persistent in their future returns, growth, and productivity, there is a fair degree of churning in the composition of star firms, especially in high-income countries. For instance, in high-income countries, 21 percent of the star firms each year were in the bottom 50 percent of ROIC in the world four years ago but that number drops to 13 percent for middle-income countries. The greater churn in the population of star firms in high-income countries points to the role that good institutions and creative destruction play in the rise of these firms.

The relation between firm markups and star status

We find that firm markups, a measure of competitive pressures at the firm level, are positively associated with greater probability of being a star firm. However, this relation is moderated by the level of intangible capital investments. In other words, the interaction of markups and intangible capital investment is negatively associated with probability of being a star firm, suggesting that for firms with high intangible capital, charging high markups is not the typical route to star status. These results are robust to alternate definitions of star status, defined both across the world and within countries, including a size criterion by defining star firms as large and profitable firms, and across both tradeable and nontradeable sectors.

On the one hand, the strong association between markups and star status could reflect greater firm efficiency: low-cost firms have higher markups and higher profits. On the other hand, the high markups and profits could also be driven by the exercise of market power in an anticompetitive setting. Establishing a valid counterfactual to test whether efficiency or market power drives the high profits of star firms in a cross-country study such as this is econometrically challenging. Hence, we approach this question by performing a number of tests to assess the market power hypothesis.

First, if star firms are achieving their star status primarily by exercising market power, we should expect them to restrict investment and output. However, we show that at every level of intangible capital, star firms have higher investment (both capital expenditure and R&D investment) and output (sales/invested capital) and this holds in both high-income and middle-income countries.

Second, if high markups and profits of star firms are due to anticompetitive behavior, we would expect an increase in competition to dampen the relationship between markups and star status. To explore this, we exploit two different competitive shocks. As a first step, we use exogenous variation in industry-level (SIC 4-digit) import tariffs as a quasi-natural experiment. The softening of trade barriers substantially increases the competitive pressure faced by a firm and therefore its ability to generate high returns for its investors. We find that increases in competition either does not affect the association between markups and star status (as we see for high-income countries and for high intangible capital firms) or positively moderates the association between markups and star status (as we see for middle-income countries and for low intangible capital firms).

The role of competition

Next, we look at an explicit case of the end of monopolistic collusion with the breakup of cartels. In a difference-in-difference setting, we compare firms that were part of a cartel (treated) over a nine-year period from t-3 to t+5 where t is the year of the end of cartel, to other firms in that country, industry, and size class that were never part of cartels (control). We see that less than 10 percent of cartel firms are star firms. Treated firms have lower markups post break-up compared to the control group. Importantly, when we split the treated firms into stars and non-stars, only the non-stars show a reduction in markups post break-up of the cartel whereas the star treated firms experience a reduction in output, and investment.

Conclusion

Our results show that there is little evidence that star firms across the world attain their status by cutting output and investment or that they are protected differentially from competition than non-star firms. While star firms have higher markups than other firms, at each level of markups we find that star firms produce more and invest more than other firms. We also find no evidence that star firms are differentially exposed to competitive shocks compared to non-star firms. First, using cuts in import tariffs to identify exogenous intensification of competition, we find no evidence that the relationship between markups and star status is dampened by increase in competitive pressures. Second, we find that for those star firms that were members of cartels, there is no evidence of a decline in markups following the end of cartel membership.

Consistent with our earlier work on US firms, we see little evidence that across the world, star firms are using market power to restrict output to generate super-normal returns for their investors or that they derive their market power from collusive and anticompetitive policies. While concerns about the potential for star firms to foreclose future competition may be warranted, our findings suggest that other factors, such as greater efficiency and/or intangible capital, may drive the association between firm markups and star status.